> (e.g. put in a separate file, or given a separate
> identifier than "T" so that it's excluded from export filters if
> necessary, can be acted on to produce curved taxiway lines, etc.)
I think that that could become quite important. I have a great fear that
if we ever get round to automatic handling of curved taxiways, and to
automatic production of the right textures at taxiway intersection etc,
then we don't want to find our taxiway data full of small little
metre-long taxiways that gunges the whole place up. I'm very worried by
taking the data denoting 'real' taxiways, and adding in lots of small
ones. Especially for the AI side of things - a network with masses of
small little nodes is going to be very nice, except exceedingly large,
and, topologically, irrelevant (except for physical movement of aircraft
models).
Giles Robertson
[EMAIL PROTECTED]
> I'm not entirely sure where the acceptable number of segments / amount
> of detail trade-off will end up. Jon Stockill has done some very
> detailed UK military layouts with hundreds of segments to show all the
> dispersal areas, but I don't think he's submitted them yet. So far
the
> airports I have done haven't had more than about 100 segments, less
than
> some of the default ones. I deliberately avoided KSQL after seeing
the
> massive extra aprons on the aerial photos! Hopefully what is
acceptable
> to Robin in terms of number of segments vs. detail vs. accuracy will
> become more clear after I get some X-Plane export filters in and users
> start submitting in a format he can accept.
Well, for the purposes of getting more stuff in the default area into
CVS, and with the possibility that "aesthetic" taxiways might be
handled differently (e.g. put in a separate file, or given a separate
identifier than "T" so that it's excluded from export filters if
necessary, can be acted on to produce curved taxiway lines, etc.),
maybe it makes the most sense for me to just forego rounding
interesections right now, and instead just get the taxiways and aprons
put in place for places I work on? I can always go back and add the
curvature (since playing with KSQL, I know I can do it OK), and by
that point maybe we'll have settled on how to treat the additional
stuff?
Of course, I have no idea how to handle KPAO in either case -- its
main feature is a large circular apron for a parking area, with a
circular taxiway running its circumference.
>> Perhaps a better way of handling curved intersection corners would
>> be to generate for each intersection a set of four boolean flags
>> indicating whether or not that corner should be rounded, and let
>> TerraGear round the corners when it puts the taxiways into the
scenery?
>> But I bet that requires hard work on their part, so maybe not such a
>> good idea.
>
>
> It's a good idea, but as you say requires a lot coding in TerraGear as
> well. Better handling of taxiway/taxiway and taxiway/runway
> intersections is something that's been mulling in the back of my mind
> for ages, but realistically I haven't got a chance of finding time to
> work on it in the foreseeable future.
I understand. At this point, I think the biggest reason I wish I could
code (well, in something other than FORTRAN or the bash shell) is
because it's gotta be a bit annoying to people in a project for other
people
(like me) to come in with ideas for *them* to implement. "Wouldn't X
be cool? Why don't you go do X?"
[ Two quotes from you ]
>>>>> There's lots of room for improvement to the airport modelling. I
>>>>> guess
>>>>> that ultimately artistic orientated folk will start doing custom
>>>>> scenery
>>>>> for certain areas, rather than relying on the automated processing
>>>>> of
>>>>> compiled data as we do at the moment.
[ snip ]
> I hope
> that eventually people will start making detailled representations of
> their local areas available, in a similar manner to MSFS scenery
> designers.
Part of MSFS' success at having so many people developing aircraft
liveries and a huge quantity of landmarks, accurate airport structures,
traffic patterns and gate usage, etc., (seen e.g. in all the stuff on
http://www.avsim.com/ and http://www.flightsims.com/ ) is doubtless
based on the number of people running it. But I also wonder if
another reason is that it's fairly well facilitated. The infrastructure
is there, and it's straightforward to use it. My understanding (but
I'm not sure; haven't run any MS software since the 80s) is that the
software one uses for creating and adding such stuff comes with MSFS,
and that the procedure for how to create stuff and how to insert it
into the existing scenery/AI/etc. is well-documented. There's little
sense of "this is going to take a lot of work just to figure out what
*to* do, before actually doing it." And when you've got something
you're happy with, there are those websites to make it available to
other people, be inspired by what others have done, etc.
We have TaxiDraw, which I think is cool, and fgsd, which I haven't
really used yet so I don't have an opinion on it. And people are
referred to Blender or AC3D for 3-d model creation. But finding
these things, learning how to use them properly (with the exception
of the TaxiDraw tutorial), etc., doesn't seem so easy. It mostly
comes from Googling and mailing list archive searching and so on.
For an example, it took a fair bit of searching the mailing list
archives to get an idea of how to relate objects created in Blender
to the desired physical sizes once placed in FlightGear scenery.
And then, once people create stuff and want to make it available
to others, and see what others have done, there's no obvious venue
for doing that. Even stuff which might get integrated into
FlightGear proper (e.g. a bunch of A320s with paintjobs of a
bunch of different airlines; or still more buildings for San
Francisco, slowing down framerates for all but those with truly
fast machines) would probably still have *some* people interested
in it. Maybe it'd be worthwhile to have a site like
http://scenery.flightgear.org/ or something like that, well-publicized
to fgfs users, where people can go to dload scenery creation stuff,
read tutorials, upload creations, etc. Maybe that would help
develop more interest. Yeah, I know, another idea in search of
someone with free time (and, in this case, bandwidth!); although I
can imagine helping with this somewhat.
-c
--
Chris Metzler [EMAIL PROTECTED]
(remove "snip-me." to email)
"As a child I understood how to give; I have forgotten this grace since
I
have become civilized." - Chief Luther Standing Bear
_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
